Developer Setup
Below are step-by-step instructions on how to set up a development environment in order to be able to successfully contribute to and execute the ecosystem test scripts.
1. Clone the ecosystem-test-scripts repository
The ecosystem test scripts are hosted on the Mozilla Github and can be cloned using the method of your choice (see Github Cloning A Repository). Contributors should follow the Contributing Guidelines and Community Participation Guidelines for the repository.
2. Create a CircleCI API Token
In order to execute the circleci_scraper script, a personal CircleCI API Token is needed. To
create a token, follow the creating a personal api token CircleCI
instructions. Store the key value in a safe place.
DO NOT SHARE YOUR CIRCLECI API TOKEN
3. Copy the BigQuery and Google Sheet Service Account JSON Keys
The metric_reporter and google_sheet_uploader scripts are set up using the
ecosystem-test-eng GCP project with the metric-reporter and metric-gsheet
service accounts respectively. In order to execute these scripts, keys
for these service account, in the form of JSON files, need to be copied from the 1Password Ecosystem
Test Engineering Team Vault into the root directory of the ecosystem-test-scripts project.
4. Set up the config.ini
All settings for the ecosystem-test-scripts are defined in the config.ini file. To set up a
local config.ini file:
4.1 Make a copy of the config.ini.sample file found in the root directory of the
ecosystem-test-scripts project and rename it to config.ini
4.2 Under the [circleci_scraper] section of the file, set the token value to the CircleCI API
key created in step 2
5. Copy the latest raw data locally
By default, CircleCI has a retention policy of 30 days for artifacts and 90 days for uploaded test
results; However, we have over a years worth of data gathered for some projects. In order to produce
reports with full trend data and reduce scraping time, copy the latest raw_data from the
ETE team folder to the root directory of the ecosystem-test-scripts project.
6. Set up the python virtual environment
This project uses Poetry for dependency management in conjunction with a pyproject.toml file. While you can use virtualenv to set up the dev environment, it is recommended to use pyenv and pyenv-virtualenv, as they work nicely with Poetry. Once poetry is installed, dependencies can be installed using the following Make command from the root directory:
make install
For more information on Make commands, run:
make help